COVID-VTS: Fact Extraction and Verification on Short Video Platforms
Fuxiao Liu, Yaser Yacoob, Abhinav Shrivastava
Abstract
We introduce a new benchmark, COVID-VTS, for fact-checking multi-modal information involving short-duration videos with COVID19- focused information from both the real world and machine generation. We propose, TwtrDetective, an effective model incorporating cross-media consistency checking to detect token-level malicious tampering in different modalities, and generate explanations. Due to the scarcity of training data, we also develop an efficient and scalable approach to automatically generate misleading video posts by event manipulation or adversarial matching. We investigate several state-of-the-art models and demonstrate the superiority of TwtrDetective.
Topics & Concepts
Computer scienceSecurity tokenBenchmark (surveying)Consistency (knowledge bases)Matching (statistics)ScalabilityCoronavirus disease 2019 (COVID-19)Event (particle physics)Adversarial systemModalitiesModel checkingArtificial intelligenceMachine learningComputer securityTheoretical computer scienceDatabaseGeodesyQuantum mechanicsSociologySocial scienceMathematicsMedicineInfectious disease (medical specialty)DiseaseStatisticsPhysicsPathologyGeographyDigital Media Forensic DetectionMultimodal Machine Learning ApplicationsAnomaly Detection Techniques and Applications